Re: Various practical colvar module related questions

From: Ajasja Ljubetič (ajasja.ljubetic_at_gmail.com)
Date: Thu Jan 20 2011 - 09:12:30 CST

Dear Jerome,

Hi Ajasja, it's been a while...
>
> I really appreciate that you have taken the time (and energy) to help me
get started with ABF. The price I'm paying for doing something totally new
in our group is that there is no one to talk to (about MD).

I have some encouraging preliminary results. A 300
ns<http://www.ijs.si/ijs/dept/epr/ajasja/mails/ABF/300ns.normalMD.png>"normal"
MD in vacum. A
0.25ns <http://www.ijs.si/ijs/dept/epr/ajasja/mails/ABF/0.25ns.abf.png> ABF
simulation (this is just fullSamples*bins) and a
75ns<http://www.ijs.si/ijs/dept/epr/ajasja/mails/ABF/75ns.ABF.png> ABF
simulation. The sampling in the 75ns ABF case looks quite good. So now I'm
of to analyze the results:)

fullSamples is not the total number of samples collected. It is the
> samples collected in each bin in the preliminary, unbiased part of the
> simulation. The larger the value, the longer the delay before the bias
> starts, so the longer it takes to explore high-free-energy regions.
>
> Aaa, the name is a bit misleading. So the point is to run the simulation
for much longer then fullSamples*number_of_bins. (and the estimate of the
free energy gradient improves with the number of steps).

> To some degree, I'd recommend smaller bins and a smaller fullSamples.
> The other parameters look okay. You could try a higher wall constant.
>
> Ok. Will set the wall constant to 10.

> > Is it feasible to split the FES into smaller regions?
>
> Is that a separate question? I don't get it.
>
> Sorry, the same question (I was re phrasing it and forgot to delete this
part).

>
> > ==> P(x) =exp(-beta*Ej)/exp(A0)
>
> Another copy/paste error on my part. Forgot to change Ej to A(x).

Thank you and best regards,
Ajasja

This archive was generated by hypermail 2.1.6 : Mon Dec 31 2012 - 23:19:44 CST